[eml-dev] EML 2.0.2 changes to text leaf nodes

[Christopher Jones]

Margaret,

About searching txt:TextType nodes:

On Mar 20, 2008, at Mar20---6:15:43 PM, Margaret O'Brien wrote:
> First, one comment on Chris's original question about drawbacks to  
> mixed content instead of xs:string: I recall one possible drawback  
> in searches. If someone searches on the string "uptake rates for  
> Alnus tenuifolia" the search will fail if <title> is text instead of  
> a string because of the embedded tags. One solution is an optional  
> <complexTitle> which is txt:TextType, in addition to <title>. The  
> search on the simple <title> returns the correct doc, and the title  
> template needs to look for the complexTitle first. Apologies for not  
> remembering who to credit for this - I think it was a verbal exchange.


To me, this seems like a database implementation detail, rather than a  
schema issue.  For instance, metacat models the DOM, and so places  
child nodes as unique records from their parents.  Other XML databases  
do this differently.  I think it is the onus of the database to  
correctly interpret data typing, and search appropriately (or at least  
create indices based on data types so that they are searched correctly).

Cheers,
Chris
_________________________________________________________________
christopher jones       cjones at msi.ucsb.edu      (805) 680-5946
marine science institute  university of california, santa barbara
_________________________________________________________________